Intelligent bus station advertisement putting method and system based on big data analysis
The bus stop advertising delivery model constructed through big data analysis and machine learning solves the accuracy of traditional bus stop advertising, realizes accurate advertising matching and resource optimization, and improves advertising effectiveness.
Patent Information
- Application Number
- CN202510421283.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-07
- Publication Date
- 2025-07-18
AI Technical Summary
The lack of accurate analysis of the advertising of traditional bus stations and cannot effectively reach the target audience, resulting in waste of advertising resources and poor results.
Through big data analysis and machine learning, bus operation data, distribution data of facilities around bus stops and traffic data are collected, and an advertisement delivery model based on machine learning is built to match and automatically deliver advertising content in real time.
It achieves precise matching between ads and target audiences, improves exposure, reduces waste of advertising resources, and improves flexibility and adaptability of advertising delivery.
Smart Images
Figure CN120338889A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of advertising placement, and particularly relates to an intelligent bus stop advertising placement method and system based on big data analysis. Background Art
[0002] In the traditional bus stop advertising placement mode, advertising placement mainly relies on experience and simple market research, lacking accurate analysis of the surrounding environment, passenger flow, and passenger characteristics of the bus stop. This leads to a lack of pertinence in advertising placement, inability to effectively reach the target audience, serious waste of advertising resources, and the advertising effect is difficult to meet expectations. With the rapid development of big data technology and machine learning technology, it provides a more accurate and efficient solution for bus stop advertising placement. By collecting and analyzing a large amount of data, it is possible to deeply understand the operation of the bus stop and the behavior characteristics of passengers, providing a direction for realizing intelligent advertising placement. Summary of the Invention
[0003] In view of the above technical deficiencies, the present invention provides an intelligent bus stop advertising placement method and system based on big data analysis, which deeply analyzes the bus operation data, the distribution data of facilities around the bus stop, and the passenger flow data of the bus stop, accurately learns the characteristics of various types of data, grasps the needs of the audience at the bus stop, and improves the efficiency of bus stop advertising placement.
[0004] The present invention is realized through the following technical solutions:
[0005] There is provided an intelligent bus stop advertising placement method based on big data analysis, and the method includes the following steps:
[0006] Step S10: Collect the operation data of the bus by docking with the information system of the bus company, collect the distribution data of facilities around each bus stop through an electronic map, and collect the passenger flow data of the bus stop by setting an infrared camera at the bus stop;
[0007] Step S20: Extract the features of the various data collected, and label each type of extracted feature;
[0008] Step S30: Construct and train a bus stop advertising placement model based on machine learning according to the extracted features;
[0009] After the model training is completed, deploy it to the bus stop, collect the operation data of the bus, the distribution data of facilities around the bus stop, and the passenger flow data of the bus stop in real time, input them into the model, and the model matches the placement content of the current bus stop advertisement according to the characteristics of the currently collected real-time data, and automatically places it on the electronic display screen or billboard of the bus stop, and automatically matches the next advertisement placement content after each advertisement is played;
[0010] The data of the facilities distribution around the bus stops in step S10 are the data of the facilities distribution within a circular area with a radius of 500 meters and each bus stop as the center.
[0011] The bus stop advertising model based on machine learning constructed in step S30 also includes sub-models established according to different advertising types.
[0012] Preferably, in step S10, the operation data of the bus including the location, driving route, arrival time and departure frequency of the bus are collected by connecting with the information system of the bus company; the distribution data of facilities around each bus stop including the distribution data of commercial facilities, residential areas and office buildings are collected through electronic maps; the passenger flow data of the bus stop including the number of people entering and leaving the bus stop in different time periods and the stay time of passengers at the bus stop are collected by setting infrared cameras at the bus stop, and all the collected data are timestamped.
[0013] Preferably, the step of extracting features from the various collected data in step S20 and labeling each type of extracted features with a label includes:
[0014] Time-based feature extraction: According to the timestamps of bus operation data and bus stop passenger flow data, determine whether the corresponding date is a weekday, weekend or holiday, mark Monday to Friday as weekdays, Saturday and Sunday as weekends, and statutory holidays as holidays; convert the timestamp into date format through Python's datetime library, and then use conditional judgment statements to mark it; then extract the time period features within a day, set 7-9 o'clock as the morning peak, 9-12 o'clock as the morning flat peak, 12-14 o'clock as the noon peak, 14-17 o'clock as the afternoon flat peak, 17-19 o'clock as the evening peak, 19-22 o'clock as the evening flat peak, and 22-7 o'clock the next day as the late night flat peak. Determine the time period according to the timestamp and mark it. For example, if the hour corresponding to the timestamp is 8, it is marked as the morning peak; at the same time, count the number of bus departures, arrival time intervals and changes in passenger flow at the bus stop in each time period as the specific values of the time period features. These specific values can reflect the operation and passenger flow of bus stations in different time periods, and provide a reference for the time dimension for advertising. The labels are "time-time period-specific time period-data indicator", such as "time-time period-morning peak-passenger flow";
[0015] Geographical location-based feature extraction: Collect data on the distribution of facilities around each bus stop according to the electronic map, determine the regional type where the bus stop is located, classify it into shopping areas, business districts, residential areas, and office building areas and label them; Collect data on the distribution of facilities around each bus stop according to the electronic map, calculate the straight-line distance or actual driving distance between the bus stop and these facilities, and count the number of various types of facilities, such as how many schools and hospitals are within 1 kilometer around. These features can reflect the environment and population needs around the bus stop, with the label "Geography - Surrounding Facilities - Facility Name - Distance / Quantity", such as "Geography - Surrounding Facilities - Subway Station - Distance 500 meters";
[0016] Population attribute-based feature extraction: Collect data on the distribution of facilities around each bus stop according to the electronic map and collect the passenger flow data of the bus stop by setting infrared cameras at the bus stop to extract the population attributes of the bus stop. When there are many office buildings around the bus stop and the passenger flow is large on weekdays, it is determined that the passengers at this bus stop are mainly office workers and labeled as white-collar workers; When there are many schools around the bus stop and the passenger flow is large during the morning and evening rush hours, it is determined that the passengers at this bus stop are mainly students and teaching staff and labeled as students or teaching staff; When there are many factories or industrial parks around the bus stop, it is determined that the passengers at this bus stop are mainly industrial workers and labeled as industrial workers.
[0017] Preferably, the step of performing feature extraction on the various data collected in step S20 and using labels to label each type of extracted feature further includes:
[0018] Feature extraction combining geographical location and population attributes: Based on the regional type where the bus stop is located determined according to the data on the distribution of facilities around each bus stop collected by the electronic map, combined with the population attributes of the bus stop extracted according to the data on the distribution of facilities around each bus stop collected by the electronic map and the passenger flow data of the bus stop collected by setting infrared cameras at the bus stop, judge the consumption ability of the passengers at the bus stop and extract the consumption ability features of the passengers; When there are high-end shopping centers, luxury stores, etc. around the bus stop, it is determined that the passengers at the bus stop have a high consumption ability and are labeled as high consumption ability; When the surrounding of the bus stop is mainly composed of budget supermarkets and ordinary convenience stores, it is determined that the passengers at the bus stop have an average consumption ability and are labeled as medium consumption ability; When there is only a small grocery store around the bus stop, it is determined that the passengers at the bus stop have a low consumption ability and are labeled as low consumption ability; These population attribute features can help the advertising placement to more accurately target the target audience.
[0019] Preferably, the step of constructing and training a machine learning-based bus stop advertising placement model according to the extracted features in step S30 includes:
[0020] Data preprocessing and advertisement tagging: Organize the features extracted in step S20 into a feature matrix in a unified format. Each row represents a data sample, that is, the feature set of a bus stop at a specific moment, and each column represents a dimension, such as "time - time period - morning rush hour - passenger flow", "geography - surrounding facilities - subway station - distance of 500 meters", etc. Ensure that the data types in the feature matrix are consistent, and perform standardization or normalization on numerical features to eliminate the impact of dimension differences on model training; Classify the advertisements and set corresponding tags, including catering advertisements, electronic product advertisements, real estate advertisements, education and learning advertisements, and medical and health advertisements;
[0021] Model construction: Adopt a network architecture based on random forest, and construct sub - models for different advertisement types, and fine - tune according to the characteristics of each advertisement type, including the initial setting of hyperparameters such as the number of decision trees, maximum depth, and minimum sample split number, to meet the needs of different advertisement placement scenarios; The number of decision trees is initially set to 100 to 200, the maximum depth is set at 5 to 10 layers to prevent overfitting, and the minimum sample split number is set to 5 to 10 to ensure that each internal node has sufficient data support when dividing;
[0022] Dataset division: After completing the preprocessing, obtain the feature dataset, and divide it into a training set, a validation set, and a test set according to the ratio of 8:1:1. The training set is used for model training and parameter learning, the validation set is used to evaluate the performance of the model, and the test set is used to evaluate the generalization ability and actual performance of the model after the model training is completed. When dividing the dataset, use the stratified sampling method to ensure that the proportion of various samples in each dataset is similar to the original dataset to ensure the accuracy of the evaluation results;
[0023] Model training: Use the training set data to initialize the training of the bus stop advertisement placement model. During the training process, the model continuously adjusts the hyperparameters to learn the mapping relationship between the feature matrix and the tags. Each decision tree grows based on a random subset of the training set. At each node, a part of the feature matrices are randomly selected from the feature matrix set for splitting. During the training process, after each training round, use the validation set to verify the model, and record the performance metrics of the model on the training set and the validation set in real - time, including accuracy, recall rate, and mean squared error, etc.;
[0024] Model evaluation: After the initialization training of the bus stop advertisement placement model is completed, use the divided test set for evaluation, and set the evaluation metrics and the corresponding qualified thresholds for each metric, including accuracy, recall rate, and mean squared error, etc.;
[0025] Model Optimization and Determination: According to the model evaluation results, the grid search method is used to optimize the hyperparameters of the model. When all evaluation indicators are greater than or equal to the corresponding set qualified thresholds, the model optimization is completed, and the optimized bus stop advertising placement model is obtained. For example, in the grid search, a value range of a hyperparameter is defined. For example, the number of decision trees in the random forest ranges from 50 to 200, with a step size of 50; the maximum depth ranges from 3 to 10, with a step size of 2. By traversing all possible hyperparameter combinations, the hyperparameter configuration with the best performance on the validation set is found. The model can be serialized and saved as a file using the pickle library or joblib library in Python, and the model can be loaded for advertising placement decisions when needed.
[0026] Preferably, after the model in step S40 matches the content of the current bus stop advertisement according to the characteristics of the currently real-time collected data, the time and frequency of the advertisement placement are set according to the current time period and the collected data. For example, for some advertisements with high exposure requirements, they are set to be displayed frequently during specific time periods of the day (such as morning and evening rush hours); for some long-term promoted advertisements, they are evenly distributed and displayed in different time periods to achieve the best publicity effect; through the built-in monitoring system of the electronic display screen or billboard, the number of times the advertisement is placed is counted. The monitoring system has a built-in human line-of-sight capture function. When passengers in the bus stop watch the advertisement placed on the electronic display screen or billboard, the number of times the advertisement is viewed is recorded, which is used to evaluate the exposure of the placed advertisement.
[0027] In addition, to achieve the above object, the present invention also proposes an intelligent bus stop advertising placement system based on big data analysis. The intelligent bus stop advertising placement system based on big data analysis includes:
[0028] Bus Stop Data Collection Module: Used to collect the operation data of buses by docking with the information system of the bus company, collect the distribution data of surrounding facilities of each bus stop through the electronic map, and collect the passenger flow data of the bus stop by setting an infrared camera at the bus stop;
[0029] Bus Stop Data Feature Extraction Module: Extract features from the collected various data, and label each type of extracted feature;
[0030] Bus Stop Advertising Placement Model Construction and Training Module: Used to construct and train a bus stop advertising placement model based on machine learning according to the extracted features;
[0031] Intelligent Bus Stop Advertising Placement Module: After the model training is completed, it is deployed to the bus stop to collect the running data of buses, the distribution data of facilities around the bus stop, and the passenger flow data of the bus stop in real time, and input them into the model. The model matches the advertising content to be placed at the current bus stop according to the characteristics of the currently collected real-time data, and automatically places it on the electronic display screen or billboard of the bus stop. After each advertisement is played, the next advertisement content is automatically re-matched;
[0032] The bus stop advertising placement model constructed and trained by the bus stop advertising placement model construction and training module also includes sub-models established according to different advertising types.
[0033] In addition, to achieve the above object, the present invention also proposes an intelligent bus stop advertising placement device based on big data analysis. The device includes: a memory, a processor, and programs such as a bus stop advertising placement algorithm based on deep learning stored on the memory and executable on the processor. The programs such as the bus stop advertising placement algorithm based on deep learning are used to implement the steps of an intelligent bus stop advertising placement method based on big data analysis as described above.
[0034] In addition, to achieve the above object, the present invention also provides a computer program product. The computer program product includes programs such as a bus stop advertising placement algorithm based on deep learning. When the programs such as the bus stop advertising placement algorithm based on deep learning are executed by a processor, they implement an intelligent bus stop advertising placement method based on big data analysis as described above.
[0035] The advantages and effects of the present invention are:
[0036] An intelligent bus stop advertising placement method and system based on big data analysis proposed by the present invention realizes the precise matching of advertisements and target audiences through the method of combining big data analysis and machine learning, improves the exposure rate of advertisements, and reduces the waste of advertising resources; at the same time, through real-time data collection and analysis, it realizes the dynamic adjustment of advertising placement strategies, and improves the flexibility and adaptability of advertising placement. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.
[0038] Figure 1 It is a flowchart of an intelligent bus stop advertising placement method based on big data analysis of the present invention.
[0039] Figure 2 This is a schematic structural diagram of an intelligent bus stop advertising placement system based on big data analysis of the present invention.
[0040] Figure 3 This is a schematic block diagram of an electronic device for intelligent bus stop advertising placement based on big data analysis of the present invention. Specific embodiments
[0041] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.
[0042] Embodiment 1, as Figure 1 shown, the present invention provides an intelligent bus stop advertising placement method and system based on big data analysis, including the following steps:
[0043] Step S10: Collect the operation data of buses by docking with the information system of the bus company, collect the distribution data of facilities around each bus stop through an electronic map, and collect the passenger flow data of the bus stop by setting an infrared camera at the bus stop.
[0044] Specifically, the distribution data of facilities around the bus stop in step S10 is the distribution data of facilities within a circular area with a radius of 500 meters centered on each bus stop.
[0045] Specifically, collecting the operation data of buses by docking with the information system of the bus company in step S10 includes the position, driving route, arrival time, and departure frequency of the buses. Collecting the distribution data of facilities around each bus stop through an electronic map includes the distribution data of commercial facilities, residential areas, and office buildings. Collecting the passenger flow data of the bus stop by setting an infrared camera at the bus stop includes the number of people entering and leaving the bus stop at different time periods and the stay time of passengers at the stop. All the collected data has a time stamp.
[0046] Step S20: Extract features from the collected various data, and label each extracted feature type.
[0047] Specifically, the steps of extracting features from the collected various data and labeling each extracted feature type in step S20 include:
[0048] Time-based feature extraction: Based on the timestamps of bus operation data and the passenger flow data at bus stops, determine whether the corresponding date is a weekday, weekend, or holiday. Mark Monday to Friday as weekdays, Saturday and Sunday as weekends, and legal holidays as holidays. Use the datetime library in Python to convert the timestamps into date formats, and then use conditional judgment statements for annotation. Then, extract the time period features within a day. Set 7-9 am as the morning rush hour, 9-12 am as the morning flat peak, 12-2 pm as the noon rush hour, 2-5 pm as the afternoon flat peak, 5-7 pm as the evening rush hour, 7-10 pm as the evening flat peak, and 10 pm to 7 am the next day as the late-night flat peak. Determine and annotate the time period based on the timestamp. For example, if the hour corresponding to the timestamp is 8, it is annotated as the morning rush hour. At the same time, count the number of bus departures, the arrival time intervals of buses, and the changes in the passenger flow at the bus stop within each time period as the specific values of the features for that time period. These specific values can reflect the operation and passenger flow conditions at the bus stop during different time periods, providing a reference in the time dimension for advertising placement. The label is "Time - Time Period - Specific Time Period - Data Index", such as "Time - Time Period - Morning Rush Hour - Passenger Flow".
[0049] Geography-based feature extraction: According to the data on the distribution of surrounding facilities of each bus stop collected from the electronic map, determine the regional type where the bus stop is located, which is classified into shopping areas, business districts, residential areas, and office building areas and annotated. According to the data on the distribution of surrounding facilities of each bus stop collected from the electronic map, calculate the straight-line distance or actual driving distance between the bus stop and these facilities, and count the number of various types of facilities, such as how many schools and hospitals are within 1 kilometer of the surrounding area. These features can reflect the environment and population needs around the bus stop. The label is "Geography - Surrounding Facilities - Facility Name - Distance / Quantity", such as "Geography - Surrounding Facilities - Subway Station - Distance 500 meters".
[0050] Population attribute-based feature extraction: According to the data on the distribution of surrounding facilities of each bus stop collected from the electronic map and the passenger flow data collected by setting up infrared cameras at the bus stops, extract the population attributes of the bus stops. When there are many office buildings around the bus stop and the passenger flow is large on weekdays, it is judged that the passengers at this bus stop are mainly office workers and annotated as white-collar workers. When there are many schools around the bus stop and the passenger flow is large during the morning rush hour and evening rush hour, it is judged that the passengers at this bus stop are mainly students and teaching staff and annotated as students or teaching staff. When there are many factories or industrial parks around the bus stop, it is judged that the passengers at this bus stop are mainly industrial workers and annotated as industrial workers.
[0051] In addition, the steps of feature extraction for the various data collected in step S20 and the use of labels to annotate each type of extracted feature also include:
[0052] Feature extraction combining geographical location and population attributes: Based on the regional type of the bus stop determined according to the distribution data of surrounding facilities collected from the electronic map, combined with the population attributes of the bus stop extracted from the distribution data of surrounding facilities collected from the electronic map and the passenger flow data collected by setting infrared cameras at the bus stop, judge the consumption ability of the passengers at the bus stop, and extract the consumption ability characteristics of the passengers; when there are high-end shopping centers, luxury stores, etc. around the bus stop, it is judged that the consumption ability of the passengers at the bus stop is relatively high and marked as high consumption ability; when the surrounding of the bus stop is mainly composed of budget supermarkets and ordinary convenience stores, it is judged that the consumption ability of the passengers at the bus stop is average and marked as medium consumption ability; when there is only a small grocery store around the bus stop, it is judged that the consumption ability of the passengers at the bus stop is relatively low and marked as low consumption ability; these population attribute characteristics can help the advertisement placement to more accurately target the audience.
[0053] Step S30: Construct and train a machine learning-based bus stop advertisement placement model according to the extracted features.
[0054] Among them, the machine learning-based bus stop advertisement placement model constructed in step S30 also includes sub-models established according to different advertisement types, including catering advertisement sub-model, electronic product advertisement sub-model, real estate advertisement sub-model, education and learning advertisement sub-model, medical and health advertisement sub-model, etc. Each sub-model sets hyperparameters according to the characteristics of the corresponding advertisement.
[0055] Specifically, the steps of constructing and training a machine learning-based bus stop advertisement placement model according to the extracted features in step S30 include:
[0056] Data preprocessing and advertisement label setting: Organize the features extracted in step S20 into a feature matrix in a unified format. Each row represents a data sample, that is, the feature set of a bus stop at a specific moment, and each column represents a dimension, such as "time - time period - morning rush hour - passenger flow", "geography - surrounding facilities - subway station - distance 500 meters", etc. Ensure that the data types in the feature matrix are consistent, and perform standardization or normalization processing on numerical features to eliminate the influence of dimension differences on model training; classify the advertisements and set corresponding labels, including catering advertisements, electronic product advertisements, real estate advertisements, education and learning advertisements, and medical and health advertisements;
[0057] Model construction: Adopt a network architecture based on random forest, construct sub-models for different types of advertisements, and fine-tune according to the characteristics of each advertisement type, including the initial setting of hyperparameters such as the number of decision trees, maximum depth, and minimum sample split number, to meet the requirements of different advertising scenarios; the number of decision trees is initially set to 100 to 200, and the maximum depth is set at 5 to 10 layers to prevent overfitting, and the minimum sample split number is set to 5 to 10 layers to ensure that each internal node has sufficient data support when dividing;
[0058] Dataset division: After completing the preprocessing, obtain the feature dataset, and divide it into a training set, a validation set, and a test set according to the ratio of 8:1:1. The training set is used for model training and parameter learning, the validation set is used to evaluate the performance of the model, and the test set is used to evaluate the generalization ability and actual performance of the model after the model training is completed. The stratified sampling method is used when dividing the dataset to ensure that the proportion of various samples in each dataset is similar to that of the original dataset to ensure the accuracy of the evaluation results;
[0059] Model training: Use the training set data to initialize the training of the bus stop advertising placement model. During the training process, the model continuously adjusts the hyperparameters to learn the mapping relationship between the feature matrix and the label. Each decision tree grows based on a random subset of the training set. At each node, a part of the feature matrices are randomly selected from the set of feature matrices for splitting. During the training process, after each training round, the validation set is used to validate the model, and the performance metrics of the model on the training set and the validation set are recorded in real time, including accuracy, recall, and mean squared error, etc.;
[0060] Model evaluation: After the initialization training of the bus stop advertising placement model is completed, use the divided test set for evaluation, and set the evaluation metrics and the corresponding qualified thresholds for each metric, including accuracy, recall, and mean squared error, etc.;
[0061] Model optimization and determination: Optimize the hyperparameters of the model using the grid search method according to the model evaluation results. When all evaluation metrics are greater than or equal to the corresponding set qualified thresholds, the model optimization is completed, and the optimized bus stop advertising placement model is obtained. For example, in the grid search, define a value range for the hyperparameters, such as the number of decision trees in the random forest ranges from 50 to 200, with a step size of 50; the maximum depth ranges from 3 to 10, with a step size of 2. By traversing all possible hyperparameter combinations, find the hyperparameter configuration with the best performance on the validation set. The model can be serialized and saved as a file using the pickle library or joblib library in Python, and the model can be loaded for advertising placement decisions when needed.
[0062] Step S40: After the model training is completed, it is deployed to the bus stop. The running data of the bus, the distribution data of the facilities around the bus stop, and the passenger flow data of the bus stop are collected in real time and input into the model. The model matches the content of the advertisement to be placed at the current bus stop according to the characteristics of the currently collected real-time data, and automatically places it on the electronic display screen or billboard of the bus stop. After each advertisement is played, the model automatically matches the content of the next advertisement to be placed.
[0063] Specifically, after the model in step S40 matches the content of the advertisement to be placed at the current bus stop according to the characteristics of the currently collected real-time data, it sets the time and frequency of the advertisement placement according to the current time period and the collected data. For example, for some advertisements with high exposure requirements, they are set to be displayed frequently during specific time periods of the day (such as morning and evening rush hours); for some advertisements with long-term promotion, they are evenly distributed and displayed in different time periods to achieve the best publicity effect; through the built-in monitoring system of the electronic display screen or billboard, the number of times the advertisement is placed is counted. The monitoring system has a built-in human line-of-sight capture function. When passengers at the bus stop watch the advertisement placed on the electronic display screen or billboard, the number of times the advertisement is viewed is recorded, which is used to evaluate the exposure of the placed advertisement.
[0064] Embodiment 2, as Figure 2 shown, the present invention also proposes an intelligent bus stop advertisement placement system based on big data analysis. The intelligent bus stop advertisement placement system based on big data analysis includes:
[0065] Bus stop data collection module: used to collect the running data of the bus by docking with the information system of the bus company, collect the distribution data of the facilities around each bus stop through the electronic map, and collect the passenger flow data of the bus stop by setting infrared cameras at the bus stop;
[0066] Bus stop data feature extraction module: extract the features of various collected data, and label each type of extracted feature;
[0067] Bus stop advertisement placement model construction and training module: used to construct and train a bus stop advertisement placement model based on machine learning according to the extracted features;
[0068] Bus stop intelligent advertisement placement module: used to be deployed to the bus stop after the model training is completed, collect the running data of the bus, the distribution data of the facilities around the bus stop, and the passenger flow data of the bus stop in real time, input them into the model, and the model matches the content of the advertisement to be placed at the current bus stop according to the characteristics of the currently collected real-time data, and automatically places it on the electronic display screen or billboard of the bus stop. After each advertisement is played, the model automatically matches the content of the next advertisement to be placed;
[0069] The bus stop advertising placement model construction and training module constructs a machine learning-based bus stop advertising placement model, which also includes sub-models established according to different advertising types.
[0070] An intelligent bus stop advertising placement system based on big data analysis provided by this application adopts an intelligent bus stop advertising placement method in Embodiment 1 above, and can solve the technical problems of low exposure and low advertising placement efficiency in traditional bus stop advertising placement methods. Compared with the prior art, the beneficial effects of the intelligent bus stop advertising placement system based on big data analysis provided by this application are the same as those of the intelligent bus stop advertising placement method based on big data analysis provided in Embodiment 1 above, and other technical features in the intelligent bus stop advertising placement system based on big data analysis are the same as the features disclosed in the method of Embodiment 1 above, and will not be elaborated here.
[0071] Embodiment 3, this application provides an intelligent bus stop advertising placement device based on big data analysis. The intelligent bus stop advertising placement device based on big data analysis includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute an intelligent bus stop advertising placement method based on big data analysis in Embodiment 1 above.
[0072] Next, refer to Figure 3 , which shows a schematic structural diagram of an intelligent bus stop advertising placement device based on big data analysis suitable for implementing Embodiment 3 of this application. The intelligent bus stop advertising placement device based on big data analysis in Embodiment 3 of this application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The shown intelligent bus stop advertising placement device is only an example and should not impose any limitation on the functions and usage scope of the embodiments of this application.
[0073] Figure 3An intelligent bus stop advertising placement device based on big data analysis as shown may include a processing system 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage system 1003 into a random access memory (RAM: Random Access Memory) 1004. In the RAM 1004, various programs and data required for the operation of an intelligent bus stop advertising placement device based on big data analysis are also stored. The processing system 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems may be connected to the I / O interface 1006: an input system 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output system 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage system 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication system 1009. The communication system 1009 can allow an intelligent bus stop advertising placement device based on big data analysis to communicate with other devices wirelessly or wiredly to exchange data. Although an intelligent bus stop advertising placement device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the shown systems. More or fewer systems may be implemented or had alternatively.
[0074] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network through the communication system, or installed from the storage system 1003, or installed from the ROM 1002. When the computer program is executed by the processing system 1001, the above functions defined in the methods of the embodiments disclosed in the present application are executed.
[0075] An intelligent bus stop advertising placement device provided by the present application adopts an intelligent bus stop advertising placement method according to Embodiment 1 above, and can solve the technical problems of low exposure and low advertising placement efficiency in the traditional bus stop advertising placement method. Compared with the prior art, the beneficial effects of the intelligent bus stop advertising placement device provided by the present application are the same as those of the intelligent bus stop advertising placement method provided by Embodiment 1 above, and other technical features in the intelligent bus stop advertising placement device are the same as those disclosed in the method of the previous embodiment, which will not be elaborated here.
[0076] Each part disclosed in the present application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in a suitable manner in any one or more embodiments or examples.
[0077] The present application also provides a computer program product, including a computer program, and when the computer program is executed by a processor, it realizes the steps of an intelligent bus stop advertising placement method as described above.
[0078] The computer program product provided by the present application can solve the technical problems of low exposure and low advertising placement efficiency in the traditional bus stop advertising placement method. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as those of the intelligent bus stop advertising placement method provided by Embodiment 1 above, which will not be elaborated here.
[0079] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these changes and modifications.
Claims
1. An intelligent bus stop advertising placement method based on big data analysis, characterized in that, The method comprises the following steps: Step S10: collecting bus operation data by connecting with the bus company's information system, collecting distribution data of facilities around each bus stop through an electronic map, and collecting passenger flow data at the bus stop by setting up infrared cameras at the bus stop; Step S20: extracting features from the collected data, and labeling each type of feature extracted; Step S30: constructing and training a bus stop advertising model based on machine learning according to the extracted features; Step S40: After the model training is completed, it is deployed to the bus station, and the operation data of the bus, the distribution data of the facilities around the bus station, and the passenger flow data of the bus station are collected in real time, and input into the model. The model matches the current bus station advertisement delivery content according to the characteristics of the current real-time collected data, and automatically delivers it to the electronic display screen or billboard of the bus station. After each advertisement is played, it automatically re-matches the next advertisement delivery content; The data of the facilities distribution around the bus stops in step S10 are the data of the facilities distribution within a circular area with a radius of 500 meters and each bus stop as the center; The bus stop advertising model based on machine learning constructed in step S30 also includes sub-models established according to different advertising types.
2. The intelligent bus stop advertising placement method based on big data analysis according to claim 1, characterized in that, In step S10, the operation data of the bus including the location, driving route, arrival time and departure frequency of the bus are collected by connecting with the information system of the bus company; the distribution data of facilities around each bus stop including the distribution data of commercial facilities, residential areas and office buildings are collected through electronic maps; the passenger flow data of the bus stop including the number of people entering and leaving the bus stop in different time periods and the length of time passengers stay at the bus stop are collected by setting infrared cameras at the bus stop; all collected data are timestamped.
3. The intelligent bus stop advertising placement method based on big data analysis according to claim 1, characterized in that, In step S20, the steps of extracting features from various collected data and labeling each type of extracted features include: Time-based feature extraction: According to the timestamps of bus operation data and bus station passenger flow data, determine whether the corresponding date is a weekday, weekend or holiday, mark Monday to Friday as weekdays, Saturday and Sunday as weekends, and statutory holidays as holidays; convert the timestamp into date format through Python's datetime library, and then use conditional judgment statements to mark it; then extract the time period features within a day, set 7-9 o'clock as the morning peak, 9-12 o'clock as the morning flat peak, 12-14 o'clock as the lunch peak, 14-17 o'clock as the afternoon flat peak, 17-19 o'clock as the evening peak, 19-22 o'clock as the evening flat peak, and 22-7 o'clock the next day as the late night flat peak. According to the timestamp, determine the time period and mark it. At the same time, count the number of bus departures, arrival time intervals and changes in bus station passenger flow in each time period as the specific values of the time period features; Geographical location-based feature extraction: Collect data on the distribution of facilities around each bus stop according to the electronic map to determine the regional type where the bus stop is located; collect data on the distribution of facilities around each bus stop according to the electronic map, calculate the straight-line distance or actual driving distance between the bus stop and these facilities, and count the number of various types of facilities; Population attribute-based feature extraction: Extract the population attributes of the bus stop according to the data on the distribution of facilities around each bus stop collected from the electronic map and the passenger flow data of the bus stop collected by setting an infrared camera at the bus stop.
4. The intelligent bus stop advertising placement method based on big data analysis according to claim 1, wherein, The steps of performing feature extraction on the various data collected in step S20 and labeling each type of extracted feature with a label further include: Feature extraction combining geographical location and population attributes: Based on the regional type where the bus stop is located determined according to the data on the distribution of facilities around each bus stop collected from the electronic map, combine the population attributes of the bus stop extracted according to the data on the distribution of facilities around each bus stop collected from the electronic map and the passenger flow data of the bus stop collected by setting an infrared camera at the bus stop, judge the consumption ability of the passengers at the bus stop, and extract the consumption ability characteristics of the passengers.
5. The intelligent bus stop advertisement placement method based on big data analysis according to claim 1, characterized in that, The steps of constructing and training a machine learning-based bus stop advertising placement model according to the extracted features in step S30 include: Data preprocessing and advertising label setting: Organize the features extracted in step S20 into a feature matrix in a unified format. Each row represents a data sample, that is, the feature set of a bus stop at a specific moment, and each column represents a dimension. Classify the advertisements and set corresponding labels, including food and beverage advertisements, electronic product advertisements, real estate advertisements, education and learning advertisements, and medical and health advertisements; Model construction: Adopt a network architecture based on random forest, construct sub-models for different advertisement types, and fine-tune according to the characteristics of each advertisement type, including the initial setting of the number of decision trees, the maximum depth, and the minimum sample split number; Dataset division: After completing the preprocessing, obtain the feature dataset, and divide it into a training set, a validation set, and a test set according to the ratio of 8:1:
1. The training set is used for model training and parameter learning, the validation set is used to evaluate the performance of the model, and the test set is used to evaluate the generalization ability and actual performance of the model after the model training is completed. The method of stratified sampling is used when dividing the dataset; Model training: Use the training set data to initialize the training of the bus stop advertising placement model. During the training process, the model continuously adjusts the hyperparameters to learn the mapping relationship between the feature matrix and the label. Each decision tree grows based on a random subset of the training set. At each node, a part of the feature matrices are randomly selected from the set of feature matrices for splitting. During the training process, after each training round, use the validation set to verify the model, and record the performance metrics of the model on the training set and the validation set in real time, including accuracy, recall rate, and mean square error; Model evaluation: After the initialization training of the bus stop advertising placement model is completed, use the divided test set for evaluation, and set the evaluation metrics and the corresponding qualified thresholds for each metric, including accuracy, recall rate, and mean square error; Model Optimization and Determination: According to the model evaluation results, the hyperparameters of the model are optimized using the grid search method. When all evaluation metrics are greater than or equal to the corresponding set qualified thresholds, the model optimization is completed, and the optimized bus stop advertising placement model is obtained.
6. The intelligent bus stop advertising placement method based on big data analysis according to claim 1, characterized in that, In step S40, after the model matches the content of the current bus stop advertisement according to the characteristics of the currently real-time collected data, the time and frequency of advertisement placement are set according to the current time period and the collected data. Through the built-in monitoring system of the electronic display screen or billboard, the number of advertisement placements is counted. The monitoring system has a built-in human line-of-sight capture function. When passengers at the bus stop watch the advertisement placed on the electronic display screen or billboard, the number of times the advertisement is viewed is recorded, which is used to evaluate the exposure of the placed advertisement.
7. An intelligent bus stop advertising delivery system based on big data analysis, characterized in that, The intelligent bus stop advertising placement system based on big data analysis includes: Bus Stop Data Collection Module: Used to collect the operation data of buses by docking with the information system of the bus company, collect the distribution data of facilities around each bus stop through the electronic map, and collect the passenger flow data of the bus stop by setting infrared cameras at the bus stop; Bus Stop Data Feature Extraction Module: Extract features from the collected various data, and label each type of extracted feature; Bus Stop Advertising Placement Model Construction and Training Module: Used to construct and train a bus stop advertising placement model based on machine learning according to the extracted features; Bus Stop Intelligent Advertising Placement Module: Used to deploy to the bus stop after the model training is completed, real-time collect the operation data of buses, the distribution data of facilities around the bus stop, and the passenger flow data of the bus stop, input them into the model, and the model matches the content of the current bus stop advertisement according to the characteristics of the currently real-time collected data, and automatically places it on the electronic display screen or billboard of the bus stop, and automatically matches the content of the next advertisement to be placed after each advertisement is played; The bus stop advertising placement model based on machine learning constructed by the bus stop advertising placement model construction and training module also includes sub-models established according to different advertisement types.
8. An intelligent bus stop advertising placement device based on big data analysis, characterized in that, The intelligent bus stop advertising placement device based on big data analysis includes: A memory, a processor, and an intelligent bus stop advertising placement program based on big data analysis stored on the memory and executable on the processor. When the intelligent bus stop advertising placement program based on big data analysis is executed by the processor, it implements the intelligent bus stop advertising placement method according to any one of claims 1 to 6.
9. A computer program product, characterized in that, The computer program product includes an intelligent bus stop advertising placement program based on big data analysis. When the intelligent bus stop advertising placement program based on big data analysis is executed by the processor, it implements the intelligent bus stop advertising placement method according to any one of claims 1 to 6.